Physical toolpath differs from simulation toolpath for Multi-Axis Swarf

[MS11880]

Hello everyone,

I am currently having trouble with the swarf strategy for multi axis milling.

The toolpath generated within Fusion as well as the toolpath and tool orientations in the simulation seem to be working fine. On the machine, the tool moves into the material and basically ruines the part in line N1568 of the nc code:

I already tried changing the lead-in values and repost the code as the crash seems to occur during the lead in of the second stepover.

When running the newly posted code on the machine, the crash happens at the exact same position.

I am new to 5-axis milling on the HAAS VF-4 with a TRT210 mounted on the table.

This is the first part to be machined using a simlutaneous strategy.

Did anyone have a similar situation in the past and at best a fix for this issue?

Running into a problem like this definetely doesn't raise my trust in the simulation...

I have attached the f3d file and the nc code to this thread for further reference.

Thanks,

Marv

[programming2C78B]

Have you properly defined and calibrated your DWO and TCPC? Is your C.O.R. where its expected to be?

[MS11880]

Yes, the calibration was done recently. 

It may be a bit hard to see in the picture - The first cut worked flawlessly. The error occurs in the transition to the second cut. 

[seth.madore]

You have your "High Feedrate Mode" set to "Preserve Rapid Movement". It's possible that you're encountering a "dog leg" rapid of some sort as it repositions. Change it to "Always use High Feed" and let's see if that eliminates the issue.

Comparing the two toolpaths (one with high feed and one with rapids):

Rapid moves are orange in this image

No rapid moves:

 

 

 

[MS11880]

As suggested, I changed the "High Feedrate Mode" to "Always use High Feed" and ran the program using the old part and damaged mill.

The machines toolpath looks better but still, during the transition between stepovers it runs into the part:

[MS11880]

Does anyone have another suggestion on what to try or on how to fix the issue?

I am currently still stuck with this issue. 

[programming2C78B]

I looked at the file and I think it's all Haas' fault here (as someone who runs one). Do other simultaneous toolpaths work on your machine properly?

[MS11880]

Thank you for taking your time to check the file.

To be honest I didn't test any other simultaneous toolpaths as I was afraid to crash again.

Did you personally have similar issues in the past?

I was thinking of giving the "Advanced Swarf" strategy a try for this part. 

[adrian8B2L2]

@MS11880   By any chance does your machine use dogleg rapid moves? 

That can often be the source of major headaches. Line 1568 is a G0 rapid move which makes me very suspicious. 

I would make sure that all linking/transition moves are done with a G1 move to ensure proper simultaneous movement.

[MS11880]

Thank you for your reply.

Setting 335 on the HAAS NGC control is already set to use linear XYZ + Rotary rapid moves: 335 - Linear Rapid Mode

I suppose the machine shouldn't use rapid dogleg moves in this configuration.

As suggested earlier by Seth, I tried switching the "High Feedrate Mode" to "Always use High Feed" and had similar (but better) results.

[programming2C78B]

I remember one time I tried to do a large countersink as a swarf and it didnt work out so I surfaced it in the end. This was earlier in my 5th axis experience so cant tell if it was me or the machine but I've done plenty of proper swarf/simultaneous toolpaths since then on our TRT210. The big thing is getting your TCPC and DWO absolutely dialed in. Is your trunnion dialed in properly? (X axis parallel, probe 0 runout, a0 flat, proper COR (double check manually).

[MS11880]

I have rechecked everything again. As it seems the TRT210 has shifted its position during the first crash.

That might also be the reason that changing "High Feedrate Mode" to "Preserve Rapid Movement" seemed to have improved the outcome but actually the tool was a couple of millimeters away from touching the part and therefore the transition that originally reached further into the part didn't seem as bad.

I have now changed to the Advanced Swarf strategy. I just ran the operation using the old mill and part and it seems to work just fine.

[programming2C78B · Accepted solution]

good thing I put my 2 cents in. We've had issues before with the X axis being out - and its not mentioned in any of the calibration or setup information.

we also added 2x toe clamps to the trunnion on its "feet" since the 2 on center dont suffice in real life.